Archived: 30 January 2006
University of Adelaide Honours thesis, November 2004.
Supervisors: Paul Coddington and Andrew Wendelborn
Recognising the importance of biodiversity to our daily lives and the pressure that human activities are placing on our living World, numerous Biodiversity Informatics initiatives have started over the years, to digitally document specimen-related information and subsequently sharing them with both researchers and the general public in an efficient manner. The Global Biodiversity Information Facility (GBIF) is an International collaboration that aims at providing free and universal access to all primary biodiversity data.
The infrastructure provided by Data Grids allows a systematic approach to the process of discovering, accessing, analysing and integration of this huge amount of biodiversity data. These data from multiple databases are geographically distributed, on heterogeneous machines and administered independently according to their own policy. Hence, collaborations like GBIF are looking towards the use of Data Grid Services in the future, such as Web Services, to replace their current system that is implemented using older web technologies through Perl and CGI. In this project, a simplified version of the GBIF network was implemented using Web Services to show that Biodiversity Informatics can benefit from Data Grid Services, while taking into account the issues relating to Biodiversity Informatics, such as metadata standards, authentication, data replication and caching. In addition, a framework using Grid Service approach was also implemented to show that it is possible for distributed database applications like GBIF to take advantage of Grid tools that were once only meant for file-based datasets.